Lignin Acidolysis Predicts Formaldehyde Generation in Pine Wood

May 2, 2017 - In wood, lignin is the major source of biogenic formaldehyde; ..... Martinez , E.; Belanche , M. I. Influence of veneer wood species on ...
0 downloads 0 Views 426KB Size
Download PDF
Research Article pubs.acs.org/journal/ascecg

Lignin Acidolysis Predicts Formaldehyde Generation in Pine Wood Guigui Wan†,‡ and Charles E. Frazier*,†,‡ †

Sustainable Biomaterials, Virginia Tech, 310 West Campus Dr., Cheatham Hall, RM 230, Blacksburg, Virginia 24061, United States Macromolecules Innovation Institute, Virginia Tech, 1075 Life Science Circle, Suite 110 (0201), Blacksburg, Virginia 24061, United States

‡

ABSTRACT: Pinus virginiana wood was heated (200 °C, 10 or 60 min) while dry or after aqueous/acid/base pretreatment in order to reveal mechanisms of formaldehyde (CH2O) generation. Consistent with prior reports, among wood structural polymers, lignin was the overwhelming source of biogenic CH2O (defined as having a carbon source in wood). Effects of wood extractives were ignored and reserved for a later report. The selection of acid catalyst strongly affected CH2O generation as predicted in the acidolysis literature of lignin model compounds and isolated lignins. Lignin methoxyl cleavage was observed but considered an unlikely source of CH2O under the experimental conditions. Alkaline pretreatments did not catalyze CH2O generation above levels observed using neutral water. Regarding wood-based composite manufacture, the implications are that lignin reactions might be manipulated during hot-pressing. Potential benefits include reduced product emissions and/or novel cross-linking strategies using biogenic CH2O. Perhaps even lignin repolymerization could be promoted for benefit, in direct opposition to biorefinery strategies for lignin removal. KEYWORDS: Lignocellulose, Wood composites, Lignin, Emissions, Regulations

■

200 °C), it has often been reported that hemicelluloses are the least thermally stable polymers in wood and that lignin is substantially more stable.13−19 However, these conclusions were based upon the determination of acid insoluble lignin (Klason lignin), which is a crude analysis that does not reveal changes in lignin structure. Preferred methods for assessing the thermal stability of in situ lignin are the lignin monomer analyses such as thioacidolysis20,21 and derivatization followed by reductive cleavage (DFRC).22 For example, thioacidolysis shows that lignin β-aryl ether bonds undergo substantial cleavage and repolymerization at heating conditions that effect little or no change in the hemicelluloses.23−25 Like most woods, Virginia pine is naturally acidic, and in situ lignin thermochemistry might be described by the classic lignin acidolysis literature. Extensive acidolysis studies on model compounds and isolated lignins indicate that β-aryl ether bonds are readily cleaved via a benzylic cation reacting in one of two pathways, C2 cleavage or C3 cleavage, as in C6C2 vs C6C3 products in Figure 1.26−32 C2 cleavage produces CH2O, a phenylacetaldehyde derivative, and a free phenol, whereas C3 cleavage produces the Hibbert’s ketones and a free phenol. The relative occurrence of these pathways depends upon the nature of the acid catalyst.29,32−34 This suggests that CH2O formation in whole wood could be manipulated by the selection of an external acid catalyst. That prediction is confirmed here, where CH2O generation in Virginia pine is studied under various heating conditions.

INTRODUCTION In the living tree or otherwise, wood contains natural, biogenic formaldehyde (CH2O), defined here as having a carbon source in wood; and heating generates much higher levels that vary according to wood moisture content, tissue type, and tree species.1−5 Biogenic CH2O levels vary dramatically for reasons not understood. For instance, Tasooji et al. demonstrated that Pinus radiata generates 2−6 times more CH2O than does Liriodendron tulipifera and Pinus virginiana when heated at 200 °C (10 min).5 Catalytic interactions involving wood extractives might explain species effects,5 but this and other hypotheses must be examined if we hope to control biogenic CH2O. Nonstructural wood-based composites are regulated to limit emissions of anthropogenic CH2O (defined here as having a fossil-fuel carbon source as in amino resin adhesives). While new amino resin technologies satisfy the latest regulations, persistent anecdotal reports cite complications attributed to biogenic CH2O. These concerns motivated an industry/ university cooperation to reveal mechanisms of CH2O generation in wood so that they could be manipulated for some benefit. All organic wood components are capable of generating CH2O, but lignin appears to be the major source.3,6 Wood extractives are not addressed here, but it is known that the extractives can directly generate CH2O,3 or they may catalyze CH2O generation,5 or they react with CH2O such that emissions are reduced.7−12 Excluding extractives, this is a study of Virginia pine (Pinus virginiana) to determine the CH2O generation potential of its structural biopolymers. Specimens were heated at 200 °C for 10 or 60 min, where the 10 min treatment crudely simulates industrial hot-press conditions. Under relatively mild heating conditions (∼25 °C − © 2017 American Chemical Society

Received: January 23, 2017 Revised: April 27, 2017 Published: May 2, 2017 4830

DOI: 10.1021/acssuschemeng.7b00264 ACS Sustainable Chem. Eng. 2017, 5, 4830−4836

Research Article

ACS Sustainable Chemistry & Engineering

Specimen Pretreatment/Heating/CH2O Measurement. Specimens were dried by cycling between vacuum (0.15 mmHg) and dry N2, three times, and then stored with dry N2/P2O5 (tree 1) or dry N2/ molecular sieves (tree 2) until reaching constant weight. Prior to heating, unextracted specimens were pretreated as follows: (1) dry control, no treatment, (2) saturated in water, (3) saturated in 10 mM acid, (4) saturated in 10 mM NaOH. Saturated specimens were subsequently adjusted to 100% moisture content. Heat treatments were conducted in sealed 50 mL serum bottles for 10 or 60 min.4 The CH2O generated during heating remained sealed in the serum bottle and was determined as described previously.4 Specimens were identified with respect to pretreatments prior to heating (dry, H2O, acid, base) and heating time, i.e., dry/10, acid/60, etc. Cellulose (Sigma-Aldrich, powder, cotton linters) and xylan (SigmaAldrich, beechwood, ≥ 90%) were pretreated as above, subjected to heating and CH2O determination as above, using 100 mg specimens. Compositional Analysis. Specimens were Soxhlet extracted (95% ethanol, 48 h), ground, sieved to retain the >40 mesh fraction, vacuum-dried, and subjected to duplicate compositional analysis as in NREL/TP-510-42618: Determination of Structural Carbohydrates and Lignin in Biomass. Buffer Capacity. Unextracted tree 1: ground and sieved to retain the 20−80 mesh fraction and then vacuum-dried over N2/P2O5. Roughly 3 g of powder was suspended in 300 mL of NaCl (1 mM) solution for 30 min with continuous stirring under N2 and then titrated with NaOH standard solution (25 mM) to reach pH 10 for acidic buffering capacity or titrated with HCl standard solution (25 mM) to reach pH 3 for alkaline buffering capacity. Analyses conducted in triplicate. Duplicate analyses of unextracted tree 2 specimens (0.2 g mature tissue) were conducted similarly using particles >40 mesh and titrated with standard solutions with concentration 20 mM in 100 mL saline. All buffering measurements were corrected using blanks (saline solution without wood). Thioacidolysis. Unextracted tree 2 and tree 1 specimens (mature tissue) were subjected to heating as described above, and control specimens (no heating), were Soxhlet extracted (95% ethanol, 48 h), air-dried, ground and sieved to retain >40 mesh fraction, and then vacuum-dried as above. Roughly 10 mg of extracted wood powder reacted with 10 mL of thioacidolysis reagent (dioxane:ethanthiol:BF3 etherate = 43.75:5:1.25, v:v:v) in a 20 mL screw-capped test tube (under dry N2, 100 °C, 4 h) with continuous stirring, as described by Lapierre et al.20,21 Thioacidolysis yields reported below are only of the common C6C3 reaction products. Methoxyl Content Determination. Tree 2 specimens (mature tissue) were Soxhlet extracted (95% ethanol, 48 h), ground and sieved to retain >40 mesh fraction, and subjected to duplicate analysis as in ASTM D1166-84: Standard Test Method for Methoxyl Groups in Wood and Related Materials.

Figure 1. Pathways for acid-catalyzed cleavage of lignin β-aryl ether linkages, exemplified in a guaiacyl lignin.26−32

■

MATERIALS AND METHODS

Two healthy Virginia pine (Pinus virginiana) trees, 70−80 years old, were felled in a natural, wooded region east of Blacksburg, VA, U.S.A., (Nellies Cave area). Stem sections (∼15 cm thick) were cut from locations that were 1.0−1.3 m above ground. Tree 1: fresh sections were air-dried for 7 days, immersed in water for 3 days, debarked, and processed into flakes (70 mm × 100 mm × 0.5 mm, tan. × long. × rad.). Prior to flaking, the first 10 years of juvenile tissue was excluded. Flakes were stored in the open laboratory at room temperature for 5 months. Prior to aqueous pretreatment/ heating, flakes were razor cut into smaller pieces (∼0.5 mm × 0.5 mm × 0.5 mm). Tree 2: fresh sections were processed into flakes as above, but immediately after felling with no water immersion and no open-air storage; juvenile and mature tissues were combined and randomized. Specimens were stored at −80 °C. Also obtained from tree 2 were six increment cores (5 mm dia) at ∼1.4 m above ground from a 100 cm2 area. They were transported to the laboratory and razor cut into ∼1 mm thick disks; juvenile (first 10 rings from pith) and mature (last 60−70 rings) tissue was isolated and stored at −80 °C. Sampling, processing, and storage of all tree 2 specimens were conducted as rapidly as possible in order to minimize atmospheric exposure; from harvest to final storage, 12 h elapsed. Prior to aqueous pretreatment/ heating, specimens were razor cut into small pieces (∼0.5 mm × 0.5 mm × 0.5 mm).

■

RESULTS AND DISCUSSION CH2O Generation in Whole Tissue. When tree 1 specimens were heated at 200 °C (60 min) in the dry state (no aqueous pretreatments), CH2O generation correlated with major thermochemical changes in lignin and minor changes in the polysaccharides, Table 1. CH2O generation was about 30 times greater than previously reported for similar specimens

Table 1. CH2O Generated in Dry, Tree 1 Specimens Due to Heating (200°C, 60 min) and Corresponding Compositional Analysis and Thioacidolysis Yield of Experimental and Control Specimensa compositional analysis (% dry wood mass) lignin

control dry/60 a

glu

xyl

gal

ara

man

sol

insol

thioacid. yield μmol/g wood

measured CH2O μmol/g wood

44.4 (0.5) 44.2 (1.1)

6.8 (0.0) 6.6 (0.1)

4.2 (0.2) 4.1 (0.1)

1.1 (0.1) 0.9 (0.0)

12.7 (0.0) 12.6 (0.2)

1.5 (0.0) 1.7 (0.0)

26.6 (0.0) 26.9 (0.0)

287.5 (13.4) 33.5 (4.2)

0.4* (0.1) 44.3 (6.3)

Carbohydrate data refers to monosaccharides. Number of measurements, n = 3, except for ∗ which is n = 7 (standard deviation). 4831

DOI: 10.1021/acssuschemeng.7b00264 ACS Sustainable Chem. Eng. 2017, 5, 4830−4836

Research Article

ACS Sustainable Chemistry & Engineering

Table 2. CH2O Generated in Tree 2 Specimens Due to 200°C Heating (10 or 60 min) under Various Conditions and Corresponding Compositional Analysis, Methoxyl Content, and Thioacidolysis Yield (C6C3 products) of Experimental and Control Specimensa based on dry wood massb % sugar glu control dry/10 H2O/10 acid/10d dry/60 H2O/60 acid/60d

42.8 42.7 41.3 40.2 42.5 39.8 34.1

(0.1) (0.1) (0.0) (0.1) (1.1) (0.2) (0.1)

xyl 6.1 6.2 4.5 4.0 6.0 3.5 2.0

(0.1) (0.6) (0.0) (0.1) (0.1) (0.0) (0.0)

% lignin

gal 2.6 2.7 1.9 1.5 2.5 1.4 1.1

(0.0) (0.3) (0.1) (0.0) (0.1) (0.0) (0.0)

ara 1.4 1.3 0.7 0.6 1.1 0.5 0.5

(0.0) (0.0) (0.1) (0.1) (0.0) (0.0) (0.0)

man 14.9 (0.1) 14.9 (0.0) 13.6 (0.1) 12.7 (0.0) 14.8 (0.2) 9.9 (0.0) 6.1 (0.0)

sol 2.4 2.5 3.6 3.8 2.6 4.5 5.8

insol

(0.1) (0.0) (0.1) (0.1) (0.0) (0.2) (0.1)

26.0 26.0 27.2 28.0 26.3 31.8 38.8

(0.3) (0.0) (0.1) (0.0) (0.0) (0.2) (0.0)

methoxyl content μmol/g dry woodc

thioacid. yield μmol/g dry woodc

CH2O μmol/g dry woodc

1596.9 (148.4) − 1006.5 (48.4) 1032.3 (32.3) − 874.3 (93.6) 735.5 (51.6)

314.6 (5.2) − 328.2 (14.9) 100.7 (18.2) − 152.5 (3.8) 26.3 (7.0)

0.1 (0.1) 0.9 (0.2) 4.2 (1.4) 29.7 (4.6) − 42.0 (1.8) 120.1 (31.7)

a

Carbohydrate data refers to monosaccharides. Number of measurements, n = 3, except for methoxyl content and thioacid. yield which is n = 2 (standard deviation). bMature plus juvenile tissue. cMature tissue. d10 mM tosylic acid.

heated at 200 °C for only 10 min.5 Among the sugars, only minor to moderate reductions in xylose and arabinose were observed. Thioacidolysis yields decreased dramatically, while the total (soluble + insoluble) lignin content increased slightly. This indicates substantial lignin cleavage and repolymerization and, as previously reported, demonstrates that Klason lignin analysis is inappropriate for judging thermal stability in lignocellulose.23−25 As per the lignin acidolysis literature, it appears that CH2O was generated through the C2 cleavage pathway, while the polysaccharides were largely resistant to heating. This interpretation is consistent with the prior analysis of isolated lignin, cellulose, and monosaccharides.3 It is reemphasized that wood extractives are not addressed here, and the complex role played by the extractives will be reported later. In that regard, wood extractives from tree 1 were likely altered through air oxidation since tree 1 specimens were stored in the open-air for five months. Tree 2 specimens were processed to minimize aging effects and subjected to a broader and more damaging array of thermal treatmentsdry and wet conditions, with or without tosylic acid catalysiswhere CH2O generation was accompanied by thermochemical changes in both lignin and the polysaccharides, Table 2. As above, heating under dry conditions caused little or no sugar decomposition, whereas aqueous pretreatments caused significant sugar degradation that was more extensive under acid catalysis. As thermal treatments became more extreme, CH2O generation increased, yields of soluble and insoluble lignin increased, and the corresponding thioacidolysis yields decreased. However, note that H2O/10 specimens generated CH2O with no detectable reduction in thioacidolysis yield. This might suggest that incipient CH2O generation is dominated by reaction of β-5 linkages (not detected in thioacidolysis). However, this is in contradiction to model compound studies,31 and the anomaly perhaps indicates a deficiency in thioacidolysis sensitivity which yields only 75− 85% of β-aryl ethers.35,36 This might suggest that CH2O determinations could offer new, more sensitive insights into lignin transformations under mild heating. However, this assertion requires careful analysis and definitive proof of CH2 O origin, as discussed later. When reductions in thioacidolysis yield were clear, it is apparent that most β-aryl ether cleavage did not generate CH2O. As percentage loss in thioacidolysis yield, CH2O generation was about 14%, 26%, and 42%, respectively, for acid/10, H2O/60, and acid/60. This suggests that C3 cleavage was always favored, as discussed later.

Note that lignin methoxyl groups were cleaved by aqueous treatments with and without external acid catalysis to an extent far exceeding CH2O generation. The resulting methanol is not expected to form CH2O under these conditions (much milder than typically required to form CH2O).37 Demethoxylation is otherwise very interesting because it will lead to lignin catechols and potentially important post-reactions. The sugar decomposition noted in Table 2 raises the distinct possibility that polysaccharides contribute to CH2O generation under more damaging heat treatments. If so, then monosaccharide decomposition did not lead to an equimolar production of CH2O, consistent with the known propensity of monosaccharides to form furans and relatively little CH2O.38,39 CH2O Generation in Polysaccharide Models. Insight on CH2O generation by wood polysaccharides was obtained by subjecting isolated cellulose and xylan to 200 °C, 10 min heating with water, and acid pretreatment, Table 3. Note that Table 3. CH2O Generation from Isolated Cellulose, Xylan, and Tree 2 Mature Tissue Due to 200°C, 10 min Heating with H2O and Acid (TsOH) Pretreatment, and Initial Levels in Unheated Controlsa measured CH2O

controlb H2O/10 acid/10

cellulose μmol/g dry sample

xylan μmol/g dry sample

tree 2 μmol/g dry sample

0.1 (0.0) 0.3 (0.1) 1.0 (0.2)

1.0 (0.1) 0.4 (0.2) 0.8 (0.1)

0.1 (0.1) 4.2 (1.4) 29.6 (4.6)

a Number of measurements, n = 3 (standard deviation). bNumbers for unheated controls specimens subtracted from those of heated specimens.

all unheated control specimens contained CH2O (particularly high in xylan), and the CH2O yields were corrected for this initial content. Relative to simple water pretreatment (H2O/ 10), acid catalysis increased CH2O generation in xylan and cellulose 2−4 times, but acid catalysis in whole wood increased CH2O generation 7-fold. On the basis of these numbers, if CH2O yields from isolated cellulose and xylan are downwardly weighted for their respective compositions in whole wood, their total contribution to CH2O generation in whole tissue would range from 2% to 3%. Schäfer and Roffael conducted similar work using isolated cellulose and various pure monosaccharides; their findings suggested that the polysaccharides would 4832

DOI: 10.1021/acssuschemeng.7b00264 ACS Sustainable Chem. Eng. 2017, 5, 4830−4836

Research Article

ACS Sustainable Chemistry & Engineering

compound studies. In other words, pathway selection is nearly exclusive among model compounds, 98−99% C2 or C3 depending upon the acid type.32 Acid effects attributed to the size of the conjugate base (selectivity of the counterion for proton removal) have been discounted, and different reaction intermediates have instead been postulated with respect to the benzyl carbon.32 Under this scenario, the tetrahedral benzyl chloride (HCl catalysis) could directly generate CH2O but much less readily than the sp2 hybridized benzyl cation with bisulfate counteranion (HSO 4 − ) resulting from H 2 SO 4 catalysis.32 Under this hypothesis, it is not clear why H2SO4 would generate more CH2O than tosylic acid but much less than DBSA would (Figure 2). However, dimeric model compounds cannot simulate the broad range of conformations expected among in situ lignin substructures, and so we could not expect dimeric models to predict all catalytic effects in whole tissue. Nevertheless, it is clear that model compound studies indeed have substantial predictive power as related to CH2O generation in wood. Note that heteronuclear single quantum correlation (HSQC) NMR spectroscopy (not described or shown here) was applied to whole tissue specimens (H2O/10, acid/10, H20/60, and acid/60; Table 2), and neither homovanillin nor C6C2 styryl ether end groups43,44 were detected. Since these structures are associated with C2 cleavage, it appears that repolymerization occurred rapidly as Lahive et al. found using model compounds.31 Related is that thioacidolysis should detect homovanillin and C6C2 styryl ethers in the form of the C6C2 thioacetal.35,36 However, its interpretation is complicated36 by the fact that the C6C2 thioacetal is also always produced as a side reaction in β-aryl ether cleavage (5% according to35 6% in our case). In this work, the yield of the C6C2 thioacetal never corresponded to the levels of CH2O detected nor to the reduction in C6C3 thioacidolysis yield. In fact, the occurrence of C6C2 styryl ethers and homovanillin end groups can only be inferred from elevations in the ratio of C6C2 thioacetal-to-C6C345 because, owing to the side reaction mentioned, the overall C6C2 thioacetal yield decreased as β-aryl ether content declined. In our case, the C6C2 thioacetal/C6C3 ratio significantly increased for acid/10, H2O/60, and acid/60. Consequently, HSQC could not detect homovanillin or C6C2 styryl ethers, but their occurrence was inferred from thioacidolysis. A reasonable explanation is that homovanillin and C6C2 styryl ethers are consumed in repolymerization and cross-linking reactions that represent another major pathway in lignin acidolysis. An interesting side note emerges. If lignin acidolysis is the principal source of biogenic formaldehyde (it remains unproven), this mechanism could be expected to operate very slowly over the life span of the tree. Consistent with this hypothesis is that the wood in living trees contains low levels of formaldehyde.5 Under this scenario, the barrier to lignin biodegradation increases since the lignin structure changes slightly during the lifespan of the tree. Consequently, the protective features identified in the “metabolic plasticity” of lignin biosynthesis46 would be augmented by lignin’s natural propensity to undergo acidolysis, thereby confronting lignindegrading microorganisms with a slowly changing structure. Although completely speculative, this hypothesis perhaps adds yet another dimension to the beautiful complexity of lignins. Sharply contrasted to acid pretreatment was that NaOH did not promote CH2O generation any more than neutral water did. However, terminal phenolic β-aryl ether and phenolic β-5

contribute to much less than 1% of the total CH2O generated in whole wood.3 Isolated polysaccharides and monosaccharides are probably poor models of in situ polysaccharide degradation. For instance, when isolated cellulose and xylan were heated in the dry state (i.e., dry/10), the measured CH2O yields (data not shown) far exceeded the corresponding polysaccharide degradation noted in Table 2. The models of in situ polysaccharide used here, and elsewhere,3 are poor and incomplete. Nevertheless, when considering all observations, Tables 1−3, we suggest that polysaccharide degradation contributes to less than 5% of the total CH2O generated in whole tissue, and during milder heating conditions it is probably much less than 1%. Other methods are required to prove this hypothesis, and once again, be reminded that effects from wood extractives are simply omitted here. A definitive picture of CH2O generation and its origins in whole wood could feasibly be obtained through the analysis of carbon isotope ratios. Stable isotopes undergo increasing degrees of fractionation as metabolic pathways are further removed from carbon fixation.40−42 Consequently, careful measurement of carbon isotope ratios in biogenic CH2O should reveal the relative contributions of lignin, polysaccharide, and extractives. Catalysis of CH2O Generation in Whole Tissue. Among the structural polymers, lignin appears to be the overwhelming source of biogenic CH2O in lignocellulose. Consequently, we should anticipate the manipulation of CH2O generation through the application of acid or alkaline pretreatments, particularly in the case of acid treatments since the lignin acidolysis pathway (C2 vs C3) is dependent on the acid type.29,32−34 Figure 2 shows that acid treatments catalyzed

Figure 2. CH2O generation from 200 °C, 10 min heating, and aqueous/acid pretreatment (tree 1) and aqueous/alkaline pretreatment (tree 2 mature tissue; ∗ blue). DBSA = dodecylbenzenesulfonic acid. Error bars = ±1 standard deviation; number of measurements, n = 3 except for H2O (n = 10), HCl (n = 9), H2SO4 (n = 12), and tosylic acid (n = 6).

CH2O generation in whole wood and at levels increasing in the order of HCl < HBr < H2SO4. Tosylic acid and HBr were equally effective, and dodecylbenzenesulfonic acid (DBSA) was the most effective for generating CH2O. Our observations are consistent with model compound studies in that H2SO4 generated more CH2O than HCl and HBr.29,32−34 However, model compound studies show that HCl and HBr both favor C3 cleavage, with very little CH2O generation (1.5% or less).29,32 In this work, thioacidolysis was only conducted for tosylic acid-treated specimens, and recall that 14% of the lignin transformation was attributed to C2 cleavage (Table 2, acid/ 10). By analogy to tosylic acid, it appears that acid treatment of whole tissue favors C3 cleavage, but CH2O is also generated and at levels substantially greater than those observed in model 4833

DOI: 10.1021/acssuschemeng.7b00264 ACS Sustainable Chem. Eng. 2017, 5, 4830−4836

ACS Sustainable Chemistry & Engineering

■

CONCLUSIONS As demonstrated by others,3,54 hot-press conditions used to manufacture wood-based composites cause the generation of natural, biogenic CH2O from wood. In order to reveal mechanisms of CH2O generation, Virginia pine specimens were subjected to heating under dry conditions or with aqueous/acid/base pretreatments. Among the wood structural polymers, it was demonstrated that lignin was the overwhelming source of CH2O, consistent with prior reports. Wellestablished acidolysis studies of lignin model compounds and isolated lignins predict that in situ, whole wood reactions will follow the C2 or C3 cleavage pathways depending upon the nature of the acid catalyst. Observations here seem to confirm that prediction, where sulfuric acid and dodecylbenzenesulfonic promoted more CH2O generation than did HCl and HBr. Lignin methoxyl cleavage was observed but considered an unlikely source of CH2O under the experimental conditions. However, methoxyl cleavage is notable for subsequent lignin reactions that might be useful. Alkaline pretreatments did not catalyze CH2O generation beyond that observed using neutral water. Not addressed here are the complex effects of wood extractives, the subject of a subsequent report. The findings suggest that lignin reactions might be subject to manipulation during composite manufacture to minimize CH2O emissions from composite panels, promote other useful reactions, or both.

structures are a known source of CH2O emission during alkaline pulping.47,48 The alkaline reaction is initiated by phenol ionization and quinone methide formation, whereafter a retroaldol reaction releases CH2O from the gamma side chain position to form styryl ethers and stilbenes. In fact, styryl ethers were detected during the cure of alkaline resol resins with pine wood.49 Under this scenario, alkaline treatment is expected to promote CH2O generation,but perhaps at lower levels than acids that react along the entire chain length. (However, Sturgeon et al. suggest that acidolysis will occur by “unzipping” initiated at phenolic end groups.30) Since alkali had no significant effect on CH2O generation (Figure 2), perhaps phenolic end groups did not react and acidolysis was suppressed by neutralization of wood acids. In this regard, note that the alkaline and acid buffering capacities of Virginia pine specimens (trees 1 and 2) were, respectively, 0.05−0.08 mM HCl and 0.03−0.05 mM NaOH, so the wood buffer capacity was easily exceeded by all acid and alkaline pretreatments used here. Consequently, it is surprising that alkaline pretreatment did not promote CH2O generation above that seen with neutral water pretreatment.

■

Research Article

INDUSTRIAL EXPLOITATION OF LIGNIN ACIDOLYSIS

■

Model studies suggest that lignin acidolysis pathways can be manipulated in solid wood, and the findings here appear to confirm that prediction. This is a satisfying testament to the great and continuing history of acidolysis research on lignin models and isolated lignins. The wood composites industry should aspire to reap benefit from this field, which is still growing through biorefinery research. For instance, lignin acidolysis studies reveal interesting strategies to manipulate the reactions depicted in and related to Figure 1, where the general goal is to improve delignification by preventing lignin repolymerization.31,50−52 In the context of wood composite manufacture, perhaps in situ lignin reactions could be manipulated for some benefit. Perhaps the intentional promotion of cross-linking through lignin repolymerization and/or reactions of biogenic CH2O could be useful. Finally, it is clear that the saturated wood conditions and acid concentrations used here do not reflect bondline conditions in wood composite manufacture. Similarly, industrial hot-press temperatures range from 150−170 °C for nonstructural composites made with amino resins, whereas temperatures at or near 200 °C are typically used for structural materials bonded with phenolic resoles. Consequently, the heating conditions used in this work were extreme relative to industrial practice. Nevertheless, the effects observed in this work could be expected to manifest themselves during industrial manufacture, but that remains unproven. The free CH2O (including hemiformals) dissolved in wood exists in a complex, moisture-dependent equilibrium.4 None have determined the precise relationship between gaseous emissions from composite products and the free CH2O in wood, but a direct relationship appears very likely.53,54 In any case, emissions from composite products are expected to depend on more than the simple time/temperature effects studied here. Hopefully, these findings can benefit the wood-based composites industry and the sustainable use of timber resources.

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Phone: +1 540 231 8318. ORCID

Charles E. Frazier: 0000-0002-5733-4744 Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This project was supported by the Wood-Based Composites Center, a National Science Foundation Industry/University Cooperative Research Center (Award No. 1035009). Partial funding was also supplied by the Virginia Agricultural Experiment Station and the McIntire Stennis Program of the National Institute of Food and Agriculture, U.S. Department of Agriculture.

■

REFERENCES

(1) Meyer, B.; Boehme, C. Formaldehyde emission from solid wood. For. Prod. J. 1997, 47, 45. (2) Martinez, E.; Belanche, M. I. Influence of veneer wood species on plywood formaldehyde emission and content. Holz Roh Werkst 2000, 58, 31−34. (3) Schafer, M.; Roffael, E. On the formaldehyde release of wood. Holz Roh Werkst 2000, 58, 259−264. (4) Tasooji, M.; Frazier, C. E. Simple Milligram-Scale Extraction of Formaldehyde from Wood. ACS Sustainable Chem. Eng. 2016, 4, 5041−5045. (5) Tasooji, M.; Wan, G.; Lewis, G.; Wise, H.; Frazier, C. E. Biogenic formaldehyde: Content and heat generation in the wood of three tree species. ACS Sustainable Chem. Eng. 2017, 5, 4243. (6) Tjeerdsma, B. F.; Boonstra, M.; Pizzi, A.; Tekely, P.; Militz, H. Characterisation of thermally modified wood: molecular reasons for wood performance improvement. Holz Roh Werkst 1998, 56, 149− 153. (7) Kalisch, J. H. Alkali-formaldehyde pretreatment of coniferous woods in sulphite pulping. Pulp Pap. Mag. Can. 1969, 21, 55−68.

4834

DOI: 10.1021/acssuschemeng.7b00264 ACS Sustainable Chem. Eng. 2017, 5, 4830−4836

Research Article

ACS Sustainable Chemistry & Engineering

dependence of acidolysis reaction on the type of acid. J. Wood Sci. 2011, 57, 219−225. (30) Sturgeon, M. R.; Kim, S.; Lawrence, K.; Paton, R. S.; Chmely, S. C.; Nimlos, M.; Foust, T. D.; Beckham, G. T. A Mechanistic Investigation of Acid-Catalyzed Cleavage of Aryl-Ether Linkages: Implications for Lignin Depolymerization in Acidic Environments. ACS Sustainable Chem. Eng. 2014, 2, 472−485. (31) Lahive, C. W.; Deuss, P. J.; Lancefield, C. S.; Sun, Z. H.; Cordes, D. B.; Young, C. M.; Tran, F.; Slawin, A. M. Z.; de Vries, J. G.; Kamer, P. C. J.; Westwood, N. J.; Barta, K. Advanced Model Compounds for Understanding Acid-Catalyzed Lignin Depolymerization: Identification of Renewable Aromatics and a Lignin-Derived Solvent. J. Am. Chem. Soc. 2016, 138, 8900−8911. (32) Yokoyama, T. Revisiting the mechanism of β-O-4 bond cleavage during acidolysis of lignin. part 6: A review. J. Wood Chem. Technol. 2015, 35, 27−42. (33) Yasuda, S.; Adachi, K.; Terashima, N.; Ota, K. Chemical Structures of Sulfuric-Acid Lignin 0.8. Reactions of 1,2-Diaryl-1,3Propanediol and Pinoresinol with Sulfuric-Acid. Mokuzai Gakkaishi 1985, 31, 125−131. (34) Karlsson, O.; Lundquist, K.; Meuller, S.; Westlid, K.; et al. On the Acidolytic Cleavage of Arylglycerol Beta-Aryl Ethers. Acta Chem. Scand. 1988, 42, 48−51. (35) Lapierre, C.; Pollet, B.; Rolando, C. New insights into the molecular architecture of hardwood lignins by chemical degradative methods. Res. Chem. Intermed. 1995, 21, 397. (36) Pasco, M. F.; Suckling, I. D. Lignin removal during kraft pulping. An investigation by thioacidolysis. Holzforschung 1994, 48, 504−508. (37) Schmidt, R. G.; Laborie, M.-P.; Frazier, C. E. Lab-scale synthesis of isotopically labeled formaldehyde for the production of formaldehyde-based wood adhesives. Holzforschung 2000, 54, 98−100. (38) Van Dam, H. E.; Kieboom, A. P. G.; Van Bekkum, H. The conversion of fructose and glucose in acidic media: formation of hydroxymethylfurfural. Starch-Stärke 1986, 38, 95−101. (39) Antal, M. J.; Leesomboon, T.; Mok, W. S.; Richards, G. N. Mechanism of formation of 2-furaldehyde from D-xylose. Carbohydr. Res. 1991, 217, 71−85. (40) Loader, N. J.; Robertson, I.; McCarroll, D. Comparison of stable carbon isotope ratios in the whole wood, cellulose and lignin of oak tree-rings. Palaeogeogr., Palaeoclimatol., Palaeoecol. 2003, 196, 395− 407. (41) Harlow, B. A.; Marshall, J. D.; Robinson, A. P. A multi-species comparison of delta13C from whole wood, extractive-free wood and holocellulose. Tree Physiol. 2006, 26, 767−74. (42) Taylor, A. M.; Brooks, J. R.; Lachenbruch, B.; Morrell, J. J.; Voelker, S. Correlation of carbon isotope ratios in the cellulose and wood extractives of Douglas-fir. Dendrochronologia 2008, 26, 125−131. (43) Hoo, L. H.; Sarkanen, K. V.; Anderson, C. D. Formation of C6 C2-Enol Ethers in the Acid-Catalyzed Hydrolysis of ErythroVeratrylglycerol-β-(2-Methoxyphenyl) Ether. J. Wood Chem. Technol. 1983, 3, 223−243. (44) Yokoyama, T.; Matsumoto, Y. Revisiting the mechanism of β-O4 bond cleavage during acidolysis of lignin. Part 2: detailed reaction mechanism of a non-phenolic C6-C2 type model compound. J. Wood Chem. Technol. 2010, 30, 269−282. (45) Lapierre, C. Determining lignin structure by chemical degradations. Lignin and Lignans, Advances in Chemistry 2010, 11−42. (46) Ralph, J.; Lundquist, K.; Brunow, G.; Lu, F.; Kim, H.; Schatz, P. F.; Marita, J. M.; Hatfield, R. D.; Ralph, S. A.; Christensen, J. H.; Boerjan, W. Lignins: natural polymers from oxidative coupling of 4hydroxyphenyl-propanoids. Phytochem. Rev. 2004, 3, 29−60. (47) Balakshin, M. Y.; Capanema, E. A.; Chen, C. L.; Gracz, H. S. Elucidation of the structures of residual and dissolved pine kraft lignins using an HMQC NMR technique. J. Agric. Food Chem. 2003, 51, 6116−6127. (48) Dimmel, D. R.; Gellerstedt, G. G. Chemistry of Alkaline Pulping. In Lignin and Ligans: Advances in Chemistry; Heitner, C., Himmel, D. R., Schmidt, J. A., Eds.; Taylor & Francis: Boca Raton, FL, 2010; pp 349−391.

(8) Lelis, R.; Roffael, E.; Becker, G.; Zum Verhalten von Splint-und. Kernholz der Kiefer bei der Verleimung mit Diisocyanat-Klebstoffen (PMDI). Holz Zentralblatt 1992, 134, 2204−2210. (9) Sachsse, S.; Roffael, R. Untersuchungen der Schälfurnier-Eignung von in Deutschland erwachsenem Douglasienholz. Holz Roh Werkst 1993, 51, 167−176. (10) Dix, B.; Roffael, E. Einfluß der Verkernung und des Baumalters auf die Eigenschaften von Spanplatten aus Kiefernholz (Pinus sylvestris) Teil 1: Kernholz und seine Eigenschaften. Holz Roh Werkst 1997, 55, 25−33. (11) Buyuksari, U.; Ayrilmis, N.; Avci, E.; Koc, E. Evaluation of the physical, mechanical properties and formaldehyde emission of particleboard manufactured from waste stone pine (Pinus pinea L.) cones. Bioresour. Technol. 2010, 101, 255−259. (12) Hoong, Y. B.; Paridah, M. T.; Loh, Y. F.; Koh, M. P.; Luqman, C. A.; Zaidon, A. Acacia mangium Tannin as Formaldehyde Scavenger for Low Molecular Weight Phenol-Formaldehyde Resin in Bonding Tropical Plywood. J. Adhes. Sci. Technol. 2010, 24, 1653−1664. (13) Kollmann, F.; Fengel, D. Changes in the chemical composition of wood by thermal treatment. Howl als Roh- und Werkstoff 1965, 23, 461−468. (14) Fengel, D.; Wegener, G. Wood: Chemistry, Ultrastructure, Reactions; Walter de Gruyter, 1984. (15) LeVan, S.; Ross, R. J.; Winandy, J. E. Effects of Fire Retardant Chemicals on the Bending Properties of Wood at Elevated Temperatures; U.S. Department of Agriculture, Forest Service, Forest Products Laboratory: Madison, WI, USA, 1990. (16) Mburu, F.; Dumarcay, S.; Huber, F.; Petrissans, M.; Gerardin, P. Evaluation of thermally modified Grevillea robusta heartwood as an alternative to shortage of wood resource in Kenya: Characterisation of physicochemical properties and improvement of bio-resistance. Bioresour. Technol. 2007, 98, 3478−3486. (17) Kocaefe, D.; Poncsak, S.; Boluk, Y. Effect of Thermal Treatment on the Chemical Composition and Mechanical Properties of Birch and Aspen. Bioresources 2008, 3, 517−537. (18) Esteves, B. M.; Pereira, H. M. Wood Modification by Heat Treatment: A Review. Bioresources 2009, 4, 370−404. (19) Gonzalez-Pena, M. M.; Curling, S. F.; Hale, M. D. C. On the effect of heat on the chemical composition and dimensions of thermally-modified wood. Polym. Degrad. Stab. 2009, 94, 2184−2193. (20) Lapierre, C.; Monties, B.; Rolando, C. Thioacidolysis of Lignin Comparison with Acidolysis. J. Wood Chem. Technol. 1985, 5, 277− 292. (21) Rolando, C.; Monties, B.; Lapierre, C. Thioacidolysis. In Methods in Lignin Chemistry; Springer: Berlin, Heidelberg, 1992; pp 334−349. (22) Lu, F. C.; Ralph, J. Derivatization followed by reductive cleavage (DFRC method), a new method for lignin analysis: Protocol for analysis of DFRC monomers. J. Agric. Food Chem. 1997, 45, 2590− 2592. (23) Windeisen, E.; Strobel, C.; Wegener, G. Chemical changes during the production of thermo-treated beech wood. Wood Sci. Technol. 2007, 41, 523−536. (24) Assor, C.; Placet, V.; Chabbert, B.; Habrant, A.; Lapierre, C.; Pollet, B.; Perre, P. Concomitant Changes in Viscoelastic Properties and Amorphous Polymers during the Hydrothermal Treatment of Hardwood and Softwood. J. Agric. Food Chem. 2009, 57, 6830−6837. (25) Brosse, N.; El Hage, R.; Chaouch, M.; Petrissans, M.; Dumarcay, S.; Gerardin, P. Investigation of the chemical modifications of beech wood lignin during heat treatment. Polym. Degrad. Stab. 2010, 95, 1721−1726. (26) Lundquist, K.; Ericsson, L. Acid degradation of lignin. III. Formation of formaldehyde. Acta Chem. Scand. 1970, 24, 3681. (27) Lundquist, K.; Lundgren, R.; et al. Acid Degradation of Lignin 0.7. Cleavage of Ether Bonds. Acta Chem. Scand. 1972, 26, 2005. (28) Lundquist, K. Low-molecular weight lignin hydrolysis products. Applied Polymer Symposia 1976, 28, 1393−1407. (29) Imai, T.; Yokoyama, T.; Matsumoto, Y. Revisiting the mechanism of beta-O-4 bond cleavage during acidolysis of lignin IV: 4835

DOI: 10.1021/acssuschemeng.7b00264 ACS Sustainable Chem. Eng. 2017, 5, 4830−4836

Research Article

ACS Sustainable Chemistry & Engineering (49) Yelle, D. J.; Ralph, J. Characterizing phenol-formaldehyde adhesive cure chemistry within the wood cell wall. Int. J. Adhes. Adhes. 2016, 70, 26. (50) Li, J.; Gellerstedt, G. Improved lignin properties and reactivity by modifications in the autohydrolysis process of aspen wood. Ind. Crops Prod. 2008, 27, 175−181. (51) Toledano, A.; Serrano, L.; Labidi, J. Improving base catalyzed lignin depolymerization by avoiding lignin repolymerization. Fuel 2014, 116, 617−624. (52) Shuai, L.; Amiri, M. T.; Questell-Santiago, Y. M.; Heroguel, F.; Li, Y. D.; Kim, H.; Meilan, R.; Chapple, C.; Ralph, J.; Luterbacher, J. S. Formaldehyde stabilization facilitates lignin monomer production during biomass depolymerization. Science 2016, 354, 329−333. (53) Myers, G. E. Mechanisms of Formaldehyde Release from Bonded Wood Products. In Formaldehyde Release from Wood Products; Myer, B., Kottes Andrews, B. A., Reinhardt, R. M., Eds.; ACS Symposium Series 316; American Chemical Society: Washington, DC, 1986; Chapter 8, pp 87−106. 10.1021/bk-1986-0316.ch008 (54) Birkeland, M. J.; Lorenz, L.; Wescott, J. M.; Frihart, C. R. Determination of native (wood derived) formaldehyde by the desiccator method in particleboards generated during panel production. Holzforschung 2010, 64, 429−433.

4836

DOI: 10.1021/acssuschemeng.7b00264 ACS Sustainable Chem. Eng. 2017, 5, 4830−4836